1SHRED(1) User Commands SHRED(1)
2
3
4
6 shred - overwrite a file to hide its contents, and optionally delete it
7
9 shred [OPTION]... FILE...
10
12 Overwrite the specified FILE(s) repeatedly, in order to make it harder
13 for even very expensive hardware probing to recover the data.
14
15 If FILE is -, shred standard output.
16
17 Mandatory arguments to long options are mandatory for short options
18 too.
19
20 -f, --force
21 change permissions to allow writing if necessary
22
23 -n, --iterations=N
24 overwrite N times instead of the default (3)
25
26 --random-source=FILE
27 get random bytes from FILE
28
29 -s, --size=N
30 shred this many bytes (suffixes like K, M, G accepted)
31
32 -u deallocate and remove file after overwriting
33
34 --remove[=HOW]
35 like -u but give control on HOW to delete; See below
36
37 -v, --verbose
38 show progress
39
40 -x, --exact
41 do not round file sizes up to the next full block;
42
43 this is the default for non-regular files
44
45 -z, --zero
46 add a final overwrite with zeros to hide shredding
47
48 --help display this help and exit
49
50 --version
51 output version information and exit
52
53 Delete FILE(s) if --remove (-u) is specified. The default is not to
54 remove the files because it is common to operate on device files like
55 /dev/hda, and those files usually should not be removed. The optional
56 HOW parameter indicates how to remove a directory entry: 'unlink' =>
57 use a standard unlink call. 'wipe' => also first obfuscate bytes in
58 the name. 'wipesync' => also sync each obfuscated byte to disk. The
59 default mode is 'wipesync', but note it can be expensive.
60
61 CAUTION: Note that shred relies on a very important assumption: that
62 the file system overwrites data in place. This is the traditional way
63 to do things, but many modern file system designs do not satisfy this
64 assumption. The following are examples of file systems on which shred
65 is not effective, or is not guaranteed to be effective in all file sys‐
66 tem modes:
67
68 * log-structured or journaled file systems, such as those supplied with
69 AIX and Solaris (and JFS, ReiserFS, XFS, Ext3, etc.)
70
71 * file systems that write redundant data and carry on even if some
72 writes fail, such as RAID-based file systems
73
74 * file systems that make snapshots, such as Network Appliance's NFS
75 server
76
77 * file systems that cache in temporary locations, such as NFS version 3
78 clients
79
80 * compressed file systems
81
82 In the case of ext3 file systems, the above disclaimer applies (and
83 shred is thus of limited effectiveness) only in data=journal mode,
84 which journals file data in addition to just metadata. In both the
85 data=ordered (default) and data=writeback modes, shred works as usual.
86 Ext3 journaling modes can be changed by adding the data=something
87 option to the mount options for a particular file system in the
88 /etc/fstab file, as documented in the mount man page (man mount).
89
90 In addition, file system backups and remote mirrors may contain copies
91 of the file that cannot be removed, and that will allow a shredded file
92 to be recovered later.
93
95 Written by Colin Plumb.
96
98 GNU coreutils online help: <https://www.gnu.org/software/coreutils/>
99 Report shred translation bugs to <https://translationproject.org/team/>
100
102 Copyright © 2018 Free Software Foundation, Inc. License GPLv3+: GNU
103 GPL version 3 or later <https://gnu.org/licenses/gpl.html>.
104 This is free software: you are free to change and redistribute it.
105 There is NO WARRANTY, to the extent permitted by law.
106
108 Full documentation at: <https://www.gnu.org/software/coreutils/shred>
109 or available locally via: info '(coreutils) shred invocation'
110
111
112
113GNU coreutils 8.30 July 2018 SHRED(1)